All Predictors, CV-Juelich

Model Infos

## Random Forest 
## 
## 4462 samples
##   20 predictor
## 
## No pre-processing
## Resampling: Cross-Validated (4 fold) 
## Summary of sample sizes: 3346, 3346, 3347, 3347 
## Resampling results across tuning parameters:
## 
##   mtry  RMSE      Rsquared   MAE     
##    2    4.632545  0.4884201  3.387492
##    5    4.534638  0.4998935  3.312374
##    7    4.522061  0.5008224  3.296893
##   10    4.519086  0.4992388  3.294105
##   12    4.512786  0.5002000  3.289123
##   15    4.533746  0.4948408  3.305485
##   20    4.550528  0.4908901  3.311565
## 
## Tuning parameter 'splitrule' was held constant at a value of variance
## 
## Tuning parameter 'min.node.size' was held constant at a value of 5
## RMSE was used to select the optimal model using the smallest value.
## The final values used for the model were mtry = 12, splitrule = variance
##  and min.node.size = 5.

Variable Importance (Gini)

Validation on Test Set

AOA Stats

## [1] "Mean Distance in Training Data: 65020.55"
## [1] "DI threshold: 0.3289"

Standard Deviation of Predictions

Forward Feature Selection, Juelich CV

Model Infos

## Random Forest 
## 
## 4462 samples
##    9 predictor
## 
## No pre-processing
## Resampling: Cross-Validated (4 fold) 
## Summary of sample sizes: 3346, 3346, 3347, 3347 
## Resampling results:
## 
##   RMSE      Rsquared   MAE     
##   4.554247  0.4943451  3.321538
## 
## Tuning parameter 'mtry' was held constant at a value of 2
## Tuning
##  parameter 'splitrule' was held constant at a value of variance
## 
## Tuning parameter 'min.node.size' was held constant at a value of 5

Variable Importance (Gini)

Validation on Test Set

AOA Stats

## [1] "Mean Distance in Training Data: 81344.13"
## [1] "DI threshold: 0.3206"

Standard Deviation of Predictions

No Latitude and Climatic Zone, Juelich-CV

Model Infos

## Random Forest 
## 
## 4462 samples
##   18 predictor
## 
## No pre-processing
## Resampling: Cross-Validated (4 fold) 
## Summary of sample sizes: 3346, 3346, 3347, 3347 
## Resampling results across tuning parameters:
## 
##   mtry  RMSE      Rsquared   MAE     
##    2    4.843144  0.4410863  3.596375
##    5    4.755628  0.4544175  3.533292
##    7    4.753775  0.4534554  3.529250
##   10    4.756186  0.4510575  3.525966
##   12    4.761877  0.4484919  3.523451
##   15    4.756936  0.4502775  3.530103
## 
## Tuning parameter 'splitrule' was held constant at a value of variance
## 
## Tuning parameter 'min.node.size' was held constant at a value of 5
## RMSE was used to select the optimal model using the smallest value.
## The final values used for the model were mtry = 7, splitrule = variance
##  and min.node.size = 5.

Variable Importance (Gini)

Validation on Test Set

AOA Stats

## [1] "Mean Distance in Training Data: 57528.2"
## [1] "DI threshold: 0.405"

Standard Deviation of Predictions

FFS, No Latitude and Climatic Zone, Juelich-CV

Model Infos

## Random Forest 
## 
## 4462 samples
##    9 predictor
## 
## No pre-processing
## Resampling: Cross-Validated (4 fold) 
## Summary of sample sizes: 3346, 3346, 3347, 3347 
## Resampling results:
## 
##   RMSE      Rsquared   MAE     
##   4.797561  0.4417623  3.568128
## 
## Tuning parameter 'mtry' was held constant at a value of 2
## Tuning
##  parameter 'splitrule' was held constant at a value of variance
## 
## Tuning parameter 'min.node.size' was held constant at a value of 5

Variable Importance (Gini)

Validation on Test Set

AOA Stats

## [1] "Mean Distance in Training Data: 72049.12"
## [1] "DI threshold: 0.3214"

Standard Deviation of Predictions